Luspatercept (ACE-536) Increases Hemoglobin and Decreases - PowerPoint PPT Presentation

Luspatercept (ACE-536) Increases Hemoglobin and Decreases Transfusion Burden and Serum Ferritin in Adults with Beta-Thalassemia: Preliminary Results from a Phase 2 Study Antonio G Piga, MD 1 , Silverio Perrotta, MD 2 , Angela Melpignano, MD 3 , Caterina Borgna- Pignatti, MD 4 , M. Rita Gamberini 4 , Ersi Voskaridou, MD 5 , Vincenzo Caruso, MD 6 , Aldo Filosa, MD 7 , Yesim Aydinok, MD 8 , Carrie Condon 9 , Dawn M. Wilson 9 , Abderrahmane Laadem, MD 10 , Matthew L. Sherman, MD 9 and Kenneth M. Attie, MD 9 1 Turin University, Italy; 2 Second University of Naples; 3 Ospedale "A. Perrino", Brindisi; 4 University of Ferrara, Italy; 5 Laiko General Hospital, Athens, Greece; 6 Garibaldi Hospital, Catania; 7 AORN "A. Cardarelli", Naples, Italy; 8 Ege University Children's Hospital, Izmir, Turkey; 9 Acceleron Pharma, Cambridge, MA; 10 Celgene Corporation, Summit, NJ, USA.

Disclosures – Dr. Antonio Piga • Research grant from Acceleron • Consultant honoraria from Celgene • Research grant from Novartis • Research grant from ApoPharma Page 1

β -Thalassemia • β -thalassemia is an inherited anemia due to defective synthesis of the β -globin chains – α -globin inclusion bodies contribute to ineffective erythropoiesis • Most severe forms require regular RBC transfusions to manage complications • Iron overload can result in major organ damage, including heart and liver, and death • Life-long daily iron chelation therapy is often inadequate in preventing iron toxicity • There are currently no safe and effective alternatives to RBC transfusion Rund D, Rachmilewitz E, NEJM 2005 Page 2

Background: Luspatercept (ACE-536) Luspatercept Modified ECD of ActRIIB receptor Fc domain of human IgG 1 antibody RBC BFU-E CFU-E Pro E Baso E Poly E Ortho E Retic EPO Luspatercept • Ineffective erythropoiesis is characterized by elevated TGF- β superfamily ligands and Smad 2/3 signaling • Luspatercept is a recombinant fusion protein containing a modified extracellular domain (ECD) of the activin receptor type IIB (ActRIIB) • Luspatercept binds to GDF11 and other ligands, inhibits Smad 2/3 signaling, and promotes late-stage erythroid differentiation Suragani R et al., Nature Med 2014 Page 3

RAP-536* Corrects Ineffective Erythropoiesis in β -Thalassemia Mouse Model (Hbb -/- ) Increased RBC Improved RBC Morphology Decreased Liver Iron wt bthal+TBS bthal+RAP-536 # # # p< 0.001 vs wt; ** p< 0.01 vs bthal + TBS Reduced Spleen Size wt bthal+TBS bthal+RAP-536 Improved Bone Mineral Density wt bthal+TBS bthal+RAP-536 wt bthal+TBS bthal+RAP-536 *RAP-536 is the murine analog of luspatercept Suragani R et al., Blood 2014 Page 4

Study Overview Luspatercept β -Thalassemia Phase 2 Clinical Trial • A phase 2, multicenter, open-label, dose escalation study in adults with β -thalassemia • Primary efficacy endpoints: • Non-transfusion dependent (NTD)*  Hb increase of ≥ 1.5 g/dL for ≥ 2 weeks • Transfusion dependent (TD)**  Transfusion burden decrease ≥ 20% over 12 weeks • Secondary endpoints: – Safety and tolerability – PK – PD such as liver iron concentration, serum ferritin, and biomarkers of erythropoiesis • Treatment: Luspatercept administered subcutaneously every 3 weeks for 3 months: Screening Luspatercept Follow-up Period Treatment Period Period Study Week -4 BL 3 6 9 12 16 20 * NTD = <4 U/8 weeks, hemoglobin < 10 g/dL ** TD = ≥ 4 U/8 weeks confirmed over 6 months NCT01749540, EudraCT 2012-002499-15 Page 5

Study Design Luspatercept β -Thalassemia Phase 2 Clinical Trial Data from completed 3 Months Treatment cohorts presented Cohort 1 Completed 0.2 mg/kg (N=6) Active Cohort 2 0.4 mg/kg (N=6) Cohort 3 0.6 mg/kg (N=6) Cohort 4 0.8 mg/kg (N=6) Cohort 5 1.0 mg/kg (N=6) Cohort 6 1.25 mg/kg (N=6) Expansion Cohort Individually titrated dose (N=30) Patients completing base study can enroll into a 12-month extension study As of 1 Dec 2014 Page 6

Baseline Characteristics Luspatercept β -Thalassemia Phase 2 Clinical Trial All Patients N=30 Age, yr, median (range) 34.5 (20-57) Sex, male (%) 16 (53%) Splenectomy (%) 25 (83%) Non-Transfusion Dependent (NTD) N= 23 (77% ) Hemoglobin, g/dL, mean ± SD 8.3 ± 0.9 Transfusion Dependent (TD) N=7 (23%) RBC Units/12 weeks, mean ± SD 7.3 ± 1.0 Data as of 10 Oct 2014 Page 7

Safety Summary Luspatercept β -Thalassemia Phase 2 Clinical Trial • No related serious adverse events – 1 grade 3 dose-limiting toxicity (worsening lumbar spine bone pain) • 3 patients discontinued early associated with an AE – 1 each with occipital headache, ankle pain, and back pain Related Adverse Events in ≥ 5% Patients 0.2 0.4 0.6 0.8 1.0 Overall mg/kg mg/kg mg/kg mg/kg mg/kg (N = 30) Preferred Term (N = 6) (N = 6) (N = 6) (N = 6) (N = 6) n (%) Bone pain 0 0 1 3 2 6 (20.0%) Headache 0 0 1 2 2 5 (16.7%) Myalgia 0 1 2 1 0 4 (13.3%) Asthenia 0 0 0 1 2 3 ( 10.0%) 0 Influenza 0 0 0 2 2 ( 6.7%) Macule 2 0 0 0 0 2 ( 6.7%) Pain in Extremity 0 0 0 2 0 2 ( 6.7%) • No development of antidrug antibodies on treatment Data as of 10 Oct 2014 Page 8

Non-Transfusion Dependent Patients

Maximum Hemoglobin Increase in NTD Patients Luspatercept β -Thalassemia Phase 2 Clinical Trial NTD, Non-transfusion dependent Data as of 10 Oct 2014 Page 10

Sustained Hemoglobin Increase in NTD Patients Luspatercept β -Thalassemia Phase 2 Clinical Trial • Higher doses (0.8-1.0 mg/kg) produced sustained increases in hemoglobin levels 0.2-0.6 mg/kg 0.8-1.0 mg/kg (N=17) (N=6) n (%) n (%) Hb increase ≥ 1.5 g/dL 0 (0%) 3 (50%) for ≥ 2 weeks (1 ° endpoint) Mean Hb increase ≥ 1.5 g/dL 0 (0%) 2 (33%) for ≥ 9 weeks NTD, Non-transfusion dependent Data as of 10 Oct 2014 Page 11

NTD Responder Hemoglobin Luspatercept β -Thalassemia Phase 2 Clinical Trial 11 Follow-up 8 Period 7 10 6 Hemoglobin (g/dL) 5 9 4 8 3 24 year old male Splenectomized 2 7 Hemoglobin 1 0.8 mg/kg 6 0 -3 BL 3 6 9 12 16 20 Weeks NTD, Non-transfusion dependent Data as of 10 Oct 2014 Page 12

Liver Iron Concentration (LIC by MRI) in NTD Patients Luspatercept β -Thalassemia Phase 2 Clinical Trial Baseline LIC ≥ 5 mg/g dry weight (dw) (n=12) • 8/12 patients had a decrease of ≥ 1 mg/g dw 3 16 weeks in LIC (mg/g dw) Change from Baseline at 2 1 0 -1 -2 On iron chelator -3 No iron chelator -4 -5 0.2 0.4 0.6 1.0 Dose (mg/kg) NTD, Non-transfusion dependent Data as of 10 Oct 2014 Page 13

Transfusion Dependent Patients

Reduced Transfusion Burden in TD Patients Luspatercept β -Thalassemia Phase 2 Clinical Trial • 7/7 (100%) patients had >60% reduction in transfusion burden over 12 weeks Includes 2 patients with β 0 β 0 genotype (79%, 75% reduction) • 0 -20 Transfusion Burden % Change in -40 -63% -67% -67% -60 -70% -75% -79% -80 * -100% -100 0.6 0.8 1.0 mg/kg mg/kg mg/kg *Based on 8 weeks data - - - Protocol-defined threshold TD, Transfusion dependent Data as of 10 Oct 2014 Page 15

TD Responder Hemoglobin Luspatercept β -Thalassemia Phase 2 Clinical Trial 13 Units Transfused Follow-up 40 year old male Period Hemoglobin Splenectomized 12 1.0 mg/kg Hemoglobin (g/dL) 11 10 9 2 2 2 2 8 1 7 6 -9 -6 -3 BL 3 6 9 12 16 20 -9 -6 -3 BL 3 6 9 12 16 20 Weeks TD, Transfusion dependent Data as of 10 Oct 2014 Page 16

Reduced Liver Iron Concentration (MRI) in TD Patients Luspatercept β -Thalassemia Phase 2 Clinical Trial Baseline LIC ≥ 5 mg/g dw (n=5) 6 16 weeks in LIC (mg/g dw) Change from Baseline at 4 2 0 -2 -4 Iron chelation therapy -6 Baseline LIC: 6.8 7.3 6.5 21.4 12.2 Max. % Decr. Ferritin: -39.7 -27.5 -59.5 -26.5 -12.3 0.6 mg/kg 0.8 mg/kg 1.0 mg/kg Page 17 Data as of 10 Oct 2014

Healing of Leg Ulcers in 2 of 2 Patients Luspatercept β -Thalassemia Phase 2 Clinical Trial Pre-Treatment After 6 Weeks NTD patient treated at 0.4 mg/kg Pre-Treatment After 4 Weeks After 18 Weeks TD patient treated at 1.0 mg/kg NTD, Non-transfusion dependent TD, Transfusion dependent Data as of 10 Oct 2014 Page 18

Conclusions Luspatercept β -Thalassemia Phase 2 Clinical Trial • Luspatercept treatment of β -thalassemia patients for 3 months at dose levels of 0.8-1.0 mg/kg demonstrated 75% of patients met the primary efficacy endpoint – Increase in hemoglobin ≥ 1.5 g/dL for ≥ 2 weeks in 50% of NTD patients – Decrease in RBC transfusion burden > 60% in 100% of TD patients • Liver iron concentration and serum ferritin decreased in TD and NTD patients • Rapid healing of leg ulcers was observed in 2 of 2 patients • The safety profile was favorable with no related serious adverse events • These data strongly support further evaluation of luspatercept in patients with β -thalassemia Page 19